Hemochromatosis (Iron Storage Disease) in Fruit Bats

Page 1

Hemochromatosis (Iron Storage Disease) in Fruit Bats

Graham Crawshaw, Sergio Oyarzun, Eduardo Valdes, and Karrie Rose

Metropolitan Toronto Zoo, Toronto, Ontario

Six Egyptian fruit bats (Rousettus aegyptiacus) were either found dead or presented with chronic

liver disease associated with accumulations of iron (Fe) within the liver as high as 2.3% dry

weight. Two other fruit bat species, the Indian flying fox (Pteropus giganteus) and the grey-

headed flying fox (Pteropus poliocephalus), also had elevated liver Fe levels but without clinical

evidence of disease. The bats were fed chopped fruit with a powdered vitamin-mineral

supplement, and a gelatinised product containing fruit, meat, and supplements. Analysis revealed

that one of the mineral sources used, mono-dicalcium phosphate, contained 11,860 mg/kg Fe dry

matter (DM). The Fe content of the complete diet was about 400 mg/kg DM which is close to the

toxic range for some domestic animals. The bats also received heavy vitamin C supplementation

amounting to an intake of over 7500 mg/kg vitamin C. Ascorbic acid increases iron uptake and

enhances Fe toxicity. Pathological findings were similar to those reported in dietary iron overload

syndromes in other mammals including man. It is recommended that dietary iron and ascorbic

acid levels in diets for frugivorous bats be closely monitored and maintained at required levels

only.

Key words: hemosiderosis, bats, Rousettus aegyptiacus, ascorbic acid

INTRODUCTION

Hemochromatosis (HC), or iron storage disease, is a pathological condition associated with

excessive iron deposition in the tissues. It has been recognized in several wild animal species in

captivity, as well as in man and domestic animals. It is common in some types of birds, notably

mynahs and related species, and toucans (Lowenstine and Petrak, 1978; Kincaid and Stoskopf,

1987; Ward et al., 1988). Iron storage disease occurs in lemurs (Benirschke et al., 1985; Gonzales

et al., 1984), and hyraxes (Rehg et al., 1978). It has also been seen in birds of paradise in the wild

(Ensley and Osborne, 1993). Cattle and horses have been affected (Lavoie and Teuscher, 1993;

House et al., 1994) and it is a significant problem in humans (Tavill et al., 1990). No reports were

found of the condition in bats.

Iron overload syndromes in humans are grouped into two categories-hereditary (primary) and

secondary hemochromatosis. Hereditary hemochromatosis is an autosomoal recessive disorder.

Approximately 1 in 200 are homozygous, but only 1 in 5 of those will develop clinical

hemochromatosis. Secondary hemochromatosis occurs in people not carrying the HC gene who

are exposed to excess quantities of dietary, medicinal, or parenteral iron, and in patients with

anemias associated with increased hematopoiesis (Powell and Halliday, 1989; Rubin and Farber,

1994). The term hemochromatosis is usually applied to the condition in which massive deposition

of iron is associated with tissue damage in the liver and other organs (Walter, 1989).

Hemosiderosis refers to presence of iron deposits in the tissues in the absence of other

pathological change.

Page 2

Hemochromatosis was responsible for the death of six Egyptian fruit bats (Rousettus

aegyptiacus) (EFB) at the Metropolitan Toronto Zoo (MTZ) between 1992 and 1994. A review

of necropsy records of EFB at MTZ revealed a high incidence of tissue hemosiderosis although,

in the earlier cases, there was no evidence that iron deposits were directly responsible for death.

Two other fruit bat species, Indian flying foxes (Pteropusgiganteus) and grey-headedflying foxes

(Pteropuspoliocephalus), also showed a high incidence of hemosiderosis. Investigation into the

cause of the condition revealed that the bats were inadvertently receiving very high levels of

dietary iron and ascorbic acid. This report describes the clinical and pathological signs of the

disease in EFB and the source of the excessive iron intake.

METHODS

All bats were born in captivity at MTZ, and were kept in a group on public exhibit or in holding

cages. The diet was prepared and offered once daily. The bats' diet consisted of approximately

50% by weight of chopped mixed fruit (banana, grape, apple, raisin, orange, pear, etc.) to which a

vitamin and mineral powder was added. Temporary additions to the supplements included pollen

powder (1991-1992), and shrimp meal (1988-1990). The remainder of the diet was a gelatinised

product containing chopped fruit and some meat blended with a vitamin supplement (SA-37,

Rogar-STB, London, ON, Canada) and minerals, that is allowed to solidify and then cut into

pieces. In the past this had proven a reliable method of ensuring the bats received an adequately

balanced diet, although intake has been variable -bats show a definite preference for the fruit

component. Dietary ingredients and quantities are listed in Tables 1 and 2.

Blood samples were taken under isoflurane anesthesia from the prepatagial or interfemoral web

veins. Biochemical analyses were performed on a Kodak Ektachem analyser (Kodak Canada Inc.,

Scarborough, ON, Canada). Serum iron levels were measured at a commercial laboratory using a

Hitachi 717 analyser. Necropsies were performed on all bats dying in the collection. Tissues were

placed in 10% buffered formalin, processed by standard methods, and stained with hematoxylin

and eosin (H & E) and Perl's Prussian blue stain. Liver iron levels were measured post mortem on

three EFB with clinical hemochromatosis (Cases 3, 4, and 5), and in one EFB, three Indian, and

four grey-headed flying foxes with widespread hemosiderosis but without clinically apparent iron

storage disease.

RESULTS

Clinical Histories

The bats affected with HC are listed in Table 3. All were adult females. Three of the six, all hand-

raised animals, were five years old, one 13 years old, one four, and one just two years old at the

time of death. Two bats were found moribund and died shortly thereafter. With the remainder, the

initial indicator was one of weight loss. Activity and appetite were usually well-maintained until

the later stages of the disease, when separation from the group occurred. The skin, mucous

membranes, and urine became icteric as the condition progressed. The livers were palpably

enlarged, extending beyond the caudal margin of the sternum, and were firm in texture.

Radiography confirmed hepatomegaly. Ascites developed in advanced cases. In the longest

surviving case (Case 4), the liver became shrunken and nodular due to extensive fibrosis. The

course of the disease varied from three weeks to several months following recognition of the

Page 3

condition. Progressive deterioration eventually necessitated euthanasia. Case 5 progressed

rapidly, due to the development of a hepatic carcinoma. Increases in serum total and conjugated

bilirubin were noted in advanced cases, but liver-associated enzyme levels showed no significant

differences between icteric and clinically normal animals. Serum iron levels were high in affected

EFB (318-480 µg/dl). Clinically normal but siderotic bats showed levels between 150 and 302

µg/dl. Clinical biochemistry and serum iron values are given in Table 4.

Two cases were treated with several courses of the iron-chelating agent deferoxamine (Desferal,

CIBA Pharmaceuticals, Mississauga, ON, Canada) 40-60 mg/kg SC once daily for up to three

weeks.

Post-mortem Findings

In each bat the tissues were markedly icteric. The livers were enlarged and rounded with an

irregular, granular surface and a firm texture. Ascites and hydrothorax were present. In some

cases the kidneys were irregular and pitted, and the spleens congested. Microscopically, normal

hepatic architecture was disrupted, and in many areas replaced, by extensive periportal and

periacinar fibrosis. Large tracts of acellular connective tissue and black pigment granules had

replaced normal cellular architecture. Hepatocyte necrosis was evident and in some locations had

progressed to areas of coagulation necrosis. Hepatocytes and Kupffer cells adjacent to the bands

of fibrosis contained large amounts of hemosiderin. The lungs showed congestion with

neutrophils, and accumulation of hemosiderin in macrophages in the alveolar septa. Iron-laden

macrophages were seen in other tissues including the spleen, small intestine, lung, and lymph

nodes. In Case 5 large areas of the liver had been invaded by a hepatocellular neoplasm. The

neoplastic cells contained very little iron but the adjacent parenchymal cells were filled with iron

deposits.

A retrospective examination of necropsy results on EFB held at the Toronto Zoo in the since

1974 revealed that in 27/47 (57%) of adult bats examined histologically, hemosiderin was in

sufficient quantity in the tissues to warrant note in the report. Extensive hemosiderosis of the

liver was seen in three EFB which died of other causes in 1985, and the level of iron in the diet

was questioned at that time. Another case with extensive pathology but which did not show

clinical disease occurred in 1992 in an EFB housed separately, but concurrently with the

problems being seen in the group. The liver was affected in 23/27 (85%) cases and the spleen in

22/27 (81 %). Other sites included the lung (33%) and the heart, small intestine, lymph node and

stomach (3% each). No iron was detected in the pancreas or the heart. Within the liver,

hemosiderin was seen in either Kupffer cells and hepatocytes, and in most cases, both. In all

cases of extrahepatic iron accumulation, hemosiderin was present in macrophages rather than

parenchymal cells. Only in the six EFBs were hemochromatosis, or its sequelae, considered the

primary cause of death. Case 5 was distinct from the other cases in the development of

hepatocellular carcinoma with local metastases. No other cases of neoplasia were seen in the bats

with, or without, hemosiderosis.

Results of post mortem liver iron levels are listed in Table 5. All three bats with HC had very

high iron levels (mean 20,00 mg/kg dry weight). The clinically normal EFB with hemosiderosis

had a liver iron content of 11,000 mg/kg. In contrast, the levels were much lower in both species

Page 4

of flying fox (mean 3596 mg/kg). None of the latter showed clinical evidence of

hemochromatosis although post-mortem examination still indicated substantial liver iron stores.

Dietary Analysis

The original calculated analysis of the whole diet suggested an iron level of about 100 mg/kg but

this was based on a presumed iron content for the mineral supplement of only 500 mg/kg.

Biochemical analysis of the bat gel in 1991 revealed an iron level of 500 mg/kg, calcium 1.64%,

phosphorus 0.68%, zinc 86 mg/kg, and copper 6.2 mg/kg on a dry matter (DM) basis. The pollen,

which was eliminated after just a year , contained 200 mg/kg iron. Subsequent laboratory analysis

of the ingredients revealed very high iron levels in the calcium-containing supplements used in

both the gel and on the fruit portion of the diet. The principal culprit was mono-dicalcium

phosphate which contained 11,860 mg/kg iron. Lower amounts were provided by the vitamin

supplement (1497 mg/kg), by limestone powder (1206 mg/kg), and steamed bone meal (208

mg/kg). The complete bat supplement powder, added to the fruit portion of the diet to increase

the mineral content, contained 3300 mg/kg iron (Table 2). Calculated analysis of the fruit portion

of the diet, including the supplement powder , revealed an iron level of 372 mg/kg DM. Feeding

trials in 1992 demonstrated that the EFB accepted approximately 65% of the fruit portion of the

diet, but less than 20% gel. The total diet consumed therefore contained approximately 400

mg/kg iron. This would amount to a total daily intake for one EFB of approximately 5 mg iron.

Results of nutritional analysis of the ration are listed in Tables 1 and 2.

High levels of vitamin C were also added to the diet in response to the requirement of the species

for extraneous vitamin C. The vitamin C level of the gel was analysed to be 6140 mg/kg. The

powder supplement, which included 5% ascorbic acid powder, contained 32,600 mg/kg vitamin

C, providing a calculated total of 7300 mg/kg DM in the fruit portion of the diet (Table 2). The

total daily intake of vitamin C for each bat was approximately 90 mg.

DISCUSSION

Initiat evaluation of the diet suggested that the source of the iron was either an ingredient in the

bat gel or was the pollen, which had been included for about a year in 1991. Most fresh fruit

contains low levels of iron. Subsequent work showed that the high dietary iron values were due

to the excessive level of iron in the mineral supplement used both in the gel and on the fruit. This

phenomenon has been encountered previously, but in most situations has proven inconsequential,

since most livestock species have a high tolerance for iron (McDowell, 1992). Many of the

minerals used to supply the calcium and phosphorus needs of animals contain high levels of iron

-ground limestone, oyster shell, and many forms of calcium phosphate may contain at least 2000-

5000 mg/kg iron. Pigs are tolerant of iron levels up to 3000 mg/kg, cattle and poultry up to 1000

mg/kg, and sheep 500 mg/kg. It has however caused problems when these mineral sources are

included in the diets of susceptible species such as mynahs, which will develop hemochromatosis

at dietary iron levels of 200 mg/kg or less. Even some commercial primate and bird diets contain

levels of iron of 300 mg/kg or more. The daily intake of more than 5 mg iron for a 130g EFB

represents 100-200 times the recommended daily allowance for humans on a per kg basis.

Dietary iron levels are not measures of potential toxicity alone since absorption rates vary greatly.

The toxicity of high levels of dietary iron depends on its bioavailablity- i.e. the form of iron, and

Page 5

the presence of other compounds which may enhance or reduce its absorption. For instance, dogs

have fed for as long as 18 months on diets containing 1% ferric oxide, while other iron salts have

proven toxic at very low intakes (NRC, 1985). The bioavailability of iron in domestic animal

feeds varies but may be estimated to be 50-60% (McDowell, 1992), and only a portion of that

may actually be taken into the circulation. Heme iron is more readily taken up than iron from

plant or mineral sources. Lemurs accumulate large iron stores on diets containing less than 300

mg/kg iron DM. It has been suggested that lemurs in captivity might be susceptible to iron

overload as a result of genetic adaptation to a natural diet containing high levels of tannins and

other dietary ingredients which bind iron, rendering it unavailable for absorption (Spelman et al.,

1989).

Vitamin C is a major factor in dietary iron absorption (Morris, 1987). Ascorbic acid increases the

potential toxicity of iron since it chelates iron and promotes the conversion of ferric salts (Fe

+++

the usual form of iron in plant material and minerals, to ferrous salts (Fe

), the form in which it

is most easily absorbed through the cells of the duodenal mucosa. Excesses of vitamin C will

increase the amount of iron absorbed by three to five-fold (Monsen, 1982), and its absence

significantly decreases iron uptake (Morris, 1987). Ascorbic acid also enhances the toxicity of

iron at the intracellular level. The cytopathological mechanisms of iron overload are not fully

understood but it is believed that toxicity principally results from lipid peroxidation and the

formation of free-radicals, which damage cell membranes and increase the fragility of cellular

lysosomes. Lipid peroxidation was rapid when normal hepatic lysosomes were exposed to iron

salts in vitro. At pH 7.4, iron released from hemosiderin could initiate lipid peroxidation only in

the presence of ascorbate (Bacon and Brit ton, 1990; Tavill et al., 1990;). The level of vitamin C

ingested by the bats was excessive and may have resulted from overenthusiastic efforts to provide

this essential nutrient. Typical levels, even in diets for primates, contain a maximum of 3000

mg/kg DM. The bats' diet contained more than twice that.

Iron is an essential nutrient due to its role in hemoglobin and intracellular cytochrome formation.

Most iron exists in the body as protein-bound complexes such as hemoglobin. Nonheme iron is

also bound in the blood to the proteins transferrin and apoferritin. In normal individuals of most

species only 30-40% of the transferrin carries iron, the remainder being known as the latent iron-

binding capacity. The level of iron in the plasma varies within individuals on a diurnal basis.

Normal serum iron levels in man are < 180 µg/dl. Typical levels are 127 µg/dl for adult males

and 113 µg/dl for adult females. Values in animals are similar or slightly higher (Morris, 1987).

In cases of hemochromatosis, iron levels are usually > 300 µg/dl representing over 95% of total

iron-binding capacity, compared with a normal of around 30% (Morris, 1987; Tavill et al., 1990).

The levels in the EFB were very high (up to 480 µg/dl) and there was good correlation between

serum iron levels and the development of HC (Table 4). However, since all bats were on the same

diet we have no truly normal values for this species.

Up to levels of about 500 mg/kg, iron is stored in the liver equally as ferritin and insoluble

hemosiderin, an amorphous compound containing ferric hydroxide and little protein. Above

1000-2000 mg/kg of iron, hemosiderin deposition predominates (Morris, 1987; McDowell,

1992). Normal liver iron levels in man and other animals are usually between 300 and 1800

mg/kg dry weight (Morris, 1987; Tavill et al., 1990). In cases of iron overload, liver iron levels

are increased to 10,000 mg/kg and up (Powell and Halliday, 1989; Tavill et al., 1990). The levels

found in the EFB were as high, or higher, than those seen in hereditary HC in man. The levels in

Page 6

the clinically normal EFB and the flying foxes were also elevated and this was reflected in the

widespread hemosiderosis in the liver, spleen, and other organs. Results from our limited number

of cases would suggest that hepatic storage of iron in bats becomes toxic above 12,000 mg/kg. It

is possible that EFBs may be absorb more iron than the other bat species, or that, in view of their

smaller size, they were receiving proportionately more iron than the larger species.

Two adult female bats diagnosed ante-mortem were treated with chelation therapy for several

months. Serial liver biopsies were taken in an attempt to document a reduction in the level of iron

in the liver histologically, as has been done in birds (Loomis and Wright, 1993). Reductions were

noted in serum bilirubin and enzyme levels but no reduction in hepatic or serum iron levels were

seen during the course of therapy. Repeated phlebotomy is the treatment of choice for human HC

and can significantly increase life expectancy. Iron chelation therapy can increase iron loss but is

expensive and is less effective for long-term treatment (Crawford and Halliday, 1991 ). Therapy

for the bats was probably not aggressive enough and the disease was probably too far advanced to

effect a recovery at that stage. In addition, the source of the iron was not detected or corrected

until after death.

In hereditary HC, iron accumulates in the hepatocytes as well as the parenchymal cells of other

organs, particularly the pancreas, until the cellular storage capacity is exceeded and cytotoxicity

occurs. Overflow of iron deposits to the cells of the reticuloendothelial system occurs as the

condition progresses but they are not primarily affected. As hepatocellular damage continues,

widespread cirrhosis develops (Powell and Halliday, 1989; Rubin and Farber, 1994). In

secondary HC, hemosiderin accumulates in the cells of the mononuclear phagocyte system,

particularly the spleen, rather than in the parenchymal tissues, but, when severe, will spillover to

the hepatocytes causing cellular death and eventually fibrosis (Walter, 1989).

The histopathology of the disease in the bats closely resembles that described in secondary HC in

man, as well as in iron overload syndromes in hyraxes, horses, and some types of birds such as

tanagers. Although hepatocytes were affected extensively in the severe cases, there was also a

wide distribution in the cells of the reticuloendothelial system. This is in contrast to the

distribution seen in mynahs, and in hereditary HC in which hepatocytes and other parenchymal

cells are the principal targets (Randell et al., 1981; Ward et al., 1988; Rubin and Farber, 1994).

Fibrosis occurs as an end-stage development in both types of HC in man, when iron levels reach

close to 20,000 mg/kg dry weight (Bassett et al., 1986). Fibrosis was extensive in most of the bat

cases which likely reflects the long duration and high level of iron deposits. Experimental

protocols have been unsuccessful at producing fibrosis in laboratory animals, presumably

because of the inadequate degree of iron overload achieved (Bacon and Brit ton, 1990).

Approximately 15-20% of human patients with hemochromatosis develop hepatocellular

carcinoma (Rubin and Farber, 1994), and the incidence of hepatic neoplasia in lemurs with

hemosiderosis was considered disproportionately high compared with unaffected primates

(Gonzales et al., 1984). One of our six cases in EFBs developed a hepatic neoplasm.

The bats' diet contained very high levels of natural and supplemented Vitamin C. Although the

absolute iron levels were still within the range for some foods and forages, iron availability and

absorption were probably maximal. It appears that the bats were faced with high levels of iron,

rendered more toxic by high levels of vitamin C. There were no other apparent causes for

increased iron uptake. Anemia was not present- red cell indices were normal or above normal.

Page 7

There is no suggestion that any of these fruit bat species are unusually susceptible to iron toxicity

when faced with moderate levels of iron intake, in the way that genetic HC patients, and mynahs

and toucans are. It is likely, however, that they are more susceptible than many domestic

mammals. An intriguing question is why sanguinivorous bats, the vampires, are able to avoid

iron overload on a diet containing as much as 3000 mg/kg iron.

This was the second occurrence of nutritional toxicity in the fruit bats at MTZ in recent years.

Proliferative bone disease caused by fluorosis was seen in the same three species of fruit bat.

High levels of fluoride were detected in shrimp meal, and in the same mono-dicalcium phosphate

powder that contributed to the iron toxicity.

CONCLUSIONS

1. Egyptian fruit bats will develop hemochromatosis when fed high levels of dietary iron.

2. Iron absorption and toxicity was likely enhanced by excessive vitamin C intake.

3. Pathologic findings in bats resembles secondary hemochromatosis in man.

4. Toxicity may result from the use of food supplements which contain high concentrations of

iron or other minerals.

REFERENCES

Adams, C.F. NUTRITIVE VALUE OF AMERICAN FOODS IN COMMON UNITS.

Agriculture Handbook No.456. Washington, DC, USDA, 1975.

Bacon, B. R., Britton, R.S. The pathology of hepatic iron overload: a free radical mediated

process? HEPATOLOGY 11: 127-137, 1990.

Bassett, M.L, Halliday, J.W, Powell, L.W. Value of hepatic iron measurements in early

hemochromatosis and determination of the critical iron level associated with fibrosis.

HEPATOLOGY 6:24-29, 1986.

Benirschke, K., Miller, C., Ippen, R., Heldstab, A. The pathology of prosimians, especially

lemurs. ADVANCES IN VETERINARY SCIENCE AND COMPARATIVE MEDICINE 30:

167-208, 1985.

Crawford, D.H.G., Halliday, J.W. Current concepts in rational therapy for haemochromatosis.

DRUGS 41 :875-882, 1991.

Ensley, P.K., Osborn, K. Hemosiderosis in a wild caught bird of paradise and other species in

Papua New Guinea. P 27 in PROCEEDINGS OF THE AMERICAN ASSOCIATION OF ZOO

VETERINARIANS, ST LOUIS, 1993.

Gonzales, J., Benirschke, K. B., Roberts, J. , Robinson, P .T. Hemosiderosis in lemurs. ZOO

BIOLOGY 3:255-265, 1984.

Kincaid, A.L., Stoskopf, M.K. Passerine dietary iron overload syndrome. ZOO BIOLOGY 6:79-

88, 1987.

Page 8

Loomis, M.R., Wright, J.F. Treatment of iron storage disease in a Bali mynah. P 28 in

PROCEEDINGS OF THE AMERICAN ASSOCIATION OF ZOO VETERINARIANS, ST

LOUIS, 1993.

Lowenstine, L.J., Petrak, M.L. Iron pigment in the liver of birds. Pp 127-135 in THE

COMPARATIVE PATHOLOGY OF ZOO ANIMALS. R. Montali; G. Migaki, eds. Washington,

DC, Smithsonian Press, 1978. McDowell, L.R. Iron. MINERALS IN HUMAN AND ANIMAL

NUTRITION. San Diego, Academic Press, Pp 152-175, 1992.

Monsen, E. Ascorbic acid: an enhancing factor in iron absorption. Ch. 5 in NUTRITIONAL

BIOAVAILABILITY OF IRON. C. Kies, ed. Washington, DC, American Chemical Society,

1982.

Morris, E.R. Iron. Pp 79-142 in TRACE ELEMENTS IN HUMAN AND ANIMAL

NUTRITION. W. Mertz, ed. San Diego, Academic Press, 1987.

National Research Council. NUTRIENT REQUIREMENTS OF DOGS. Washington, DC,

National Academy of Sciences, 1985.

Powell, L. W., Halliday, J.W. Hemochromatosis. Pp 3203-3221 in BOCKUS

GASTROENTEROLOGY. J.E. Berk, W.S. Haubrich, M.H. Kaiser, J.L.A. Roth; F. Schaffner,

eds. Philadelphia, W.B. Saunders, 1985.

Randell, M.G., Patnaik, A.K., Gould, W .J. Hepatopathy associated with excessive iron storage in

mynah birds. JOURNAL OF THE AMERICAN VETERINARY MEDICAL ASSOCIATION

179:1214-1217,1981.

Rehg, J.E., Burek, J.D., Strandberg, J.D., Montali, R.J. Hemochromatosis in the rock hyrax. Pp

113-119 in THE COMPARATIVE PATHOLOGY OF ZOO ANIMALS. R. Montali; G. Migaki,

eds. Washington, DC, Smithsonian Press, 1978.

Rubin, E., Farber, J.L. PATHOLOGY .Philadelphia, J.B. Lippincott, 1994.

Spelman, L.H., Osborn, K.G., Anderson, M.P. 1989. Pathogenesis of hemosiderosis in lemurs:

role of dietary iron, tannin, and ascorbic acid. ZOO BIOLOGY 8:239-251.

Tavill, A.S., Sharma, B.K., Bacon, B.R. Iron and the liver: genetic hemochromatosis and other

hepatic iron overload disorders. Pp 281-305 in PROGRESS IN LIVER DISEASES, Vol IX. H.

Popper; F. Schaffner, eds. Philadelphia, W .8. Saunders, 1990.

Walter, J.B. Liver. THE PATHOLOGY OF HUMAN DISEASE. Philadelphia, Lea & Febiger,

Pp 645-646, 1989.

Ward, R.J., lancu, T.C., Henderson, G.M., Kirkwood, J.K., Peters, T.J. Hepatic iron overload in

avian species: analytical and morphological studies. AVIAN PATHOLOGY 17:451-464,1988.

Page 9

Page 10

Page 11